Bug#1058701: pm-utils: unauthorised and uncommunicated removal
Joerg Jaspert writes ("Re: pm-utils: unauthorised and uncommunicated removal"):
> It wouldn't have made the review any noticable amount harder.
Fair enough.
> I do suggest fixing those things, Ancient Standard version and debhelper
> 7 do make it *really* easy to accept "This is unmaintained".
Yes.
Thanks,
Ian.
--
Ian JacksonThese opinions are my own.
Pronouns: they/he. If I emailed you from @fyvzl.net or @evade.org.uk,
that is a private address which bypasses my fierce spamfilter.
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
On 17085 March 1977, Ian Jackson wrote: Yes, please do so. Thanks. This has now been done. I chose to *not* fix anything about the package now (not even the wrong VCS fields, for example) in order to simplify review. The diff against the version previously in sid, and currently in testing, is attached. It wouldn't have made the review any noticable amount harder. I do suggest fixing those things, Ancient Standard version and debhelper 7 do make it *really* easy to accept "This is unmaintained". -- bye, Joerg
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
Thorsten Alteholz writes ("Re: pm-utils: unauthorised and uncommunicated
removal"):
> On Thu, 21 Dec 2023, Ian Jackson wrote:
> > I intend to re-upload the last version shortly (and reopen all the
> > bug reports).
>
> Yes, please do so.
Thanks. This has now been done.
I chose to *not* fix anything about the package now (not even the
wrong VCS fields, for example) in order to simplify review.
The diff against the version previously in sid, and currently in
testing, is attached.
Regards,
Ian.
diff --git a/debian/changelog b/debian/changelog
index 9dd8325..7469392 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,10 @@
+pm-utils (1.4.1-20) unstable; urgency=medium
+
+ * No-change upload to reintroduce to sid following erroneous
+removal (#1058701).
+
+ -- Ian Jackson Fri, 22 Dec 2023 22:26:26
+
+
pm-utils (1.4.1-19) unstable; urgency=medium
* No-change upload to make myself the maintainer. I intend to perhaps
--
Ian JacksonThese opinions are my own.
Pronouns: they/he. If I emailed you from @fyvzl.net or @evade.org.uk,
that is a private address which bypasses my fierce spamfilter.
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
On Thu, 21 Dec 2023 21:57:33 -0500, Paul R. Tagliamonte wrote: Thanks for your thoughtful response. And I share your conclusion: > This specific situation seems unfortunate. I have every confidence the > maintainers involved will collaborate in a good faith effort to move > the distro forward. Cheers, gregor -- .''`. https://info.comodo.priv.at -- Debian Developer https://www.debian.org : :' : OpenPGP fingerprint D1E1 316E 93A7 60A8 104D 85FA BB3A 6801 8649 AA06 `. `' Member VIBE!AT & SPI Inc. -- Supporter Free Software Foundation Europe `- signature.asc Description: Digital Signature
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
Hi. Thanks for your nice email.
Thorsten Alteholz writes ("Re: pm-utils: unauthorised and uncommunicated
removal"):
> this is sad. The RM bug appeared on the tracker page of the package, in
> your packages overview, on the ftpmaster removals page (or on the bug
> page). It was also sent to debian-bugs-dist@lists.debian.org.
Right. But none of those places actually email the maintainer.
> Where would have been a better place to draw your attention to it?
I think ideally the process would have involved a bug against the
package. I appreciate that that's not always convenient.
Would it be possible and sensible to improve your tooling to warn you,
or require additional configuration, when you're removing a package
from sid that is still in testing ? I think that would be an unusual
case. (But that's just a guess; maybe I'm wrong about how unusual it
is.) (And apparently this is quite rare and maybe this bug isn't the
right place to discuss such an improvement.)
> Hmm, the standards version is 3.9.7, the VCS URLs point to a no longer
> existing repository, the last upload was in 2019. This looks rather like
> an abandoned package. But of course, your mileage may vary
Yes. I can see how you would think that. Updating the package to
improve the metadata etc. didn't seem a priority.
For the record, I think your action was perfectly understandable; you
normally don't need to consider whether a removal request is the
result of ill will or conflict. And, I very much appreciate all the
hard work you do with the day to day of ftpmaster. It is difficult
for me to express my thanks for that strongly enough. I understand
that a person who does a lot of work will make the occasional mistake,
too.
I should probably have said all this in my last mail.
> > I intend to re-upload the last version shortly (and reopen all the
> > bug reports).
>
> Yes, please do so.
Thanks. That'll probably happen later today.
Ian.
--
Ian JacksonThese opinions are my own.
Pronouns: they/he. If I emailed you from @fyvzl.net or @evade.org.uk,
that is a private address which bypasses my fierce spamfilter.
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
> I agree. > Removal request by maintainers are fine. > Removal requests by anyone for un-maintained packages are ok. > Removal requests by third-parties for packages with a maintainer are > a situation to take a closer look at least. Without: 1) my ftpteam hat on 2) any specific reference to this individual situation 3) this being a direct reply to the above (this isn't about your email gregor, I totally understand what you're saying; and I think if this was on a package that had recent uploads it would have triggered some additional scrutiny, which means we would do most of the above as status-quo) I will note a removal being done against a maintainer's wishes is incredibly rare. I think among the thousands of removals I've watched, I can not remember another instance of a removal being done to a package where the maintainer disagreed. This is all to say, adding to the layers of burden when processing a removal that was properly filed, and passed the sniff test (RoQA is inherently not by the maintainer, and the package /looked/ unmaintained, even if it wasn't in reality) for an issue that has only happened -- as far as I know -- once in modern history, would be a tough outcome. I don't see people weaponizing the removal process, and it seems like a pretty straight-forward thing to address if folks started to maliciously file removals. This specific situation seems unfortunate. I have every confidence the maintainers involved will collaborate in a good faith effort to move the distro forward. If that means re-uploading pm-utils, a fast-track trip through NEW isn't hard. I don't think this would impact any of our users in any meaningful sense (package is still installed, plus it's sid) -- I don't think a change to the process is warranted. This seems like a social problem folks ought to work through. Paul
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
On Fri, 22 Dec 2023 02:04:52 +, Thorsten Glaser wrote: > I’ll argue for two things here: > • for a removal, if the requester is not the maintainer, check > back with the maintainer. Just always do that. I agree. Removal request by maintainers are fine. Removal requests by anyone for un-maintained packages are ok. Removal requests by third-parties for packages with a maintainer are a situation to take a closer look at least. Cheers, gregor -- .''`. https://info.comodo.priv.at -- Debian Developer https://www.debian.org : :' : OpenPGP fingerprint D1E1 316E 93A7 60A8 104D 85FA BB3A 6801 8649 AA06 `. `' Member VIBE!AT & SPI Inc. -- Supporter Free Software Foundation Europe `- signature.asc Description: Digital Signature
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
Thorsten Alteholz dixit: > Where would have been a better place to draw your attention to it? Not Ian, but the *extremely* obvious and expected place would be eMail to the maintainer. (Actually, probably: an eMail to everyone who would have gotten an eMail if a bug against the package itself had been reported.) All those ways you noticed, save d-bugs-dist (which is high-volume and generic and therefore not a good way) are polling, not really suitable for notification. I’ll argue for two things here: • for a removal, if the requester is not the maintainer, check back with the maintainer. Just always do that. • perhaps a chance to extend debbugs to copy ftp.d.o bugs that affect a package, no matter what, to the bugreport recipients that that package would have gotten. bye, //mirabilos -- 21:49⎜ I have a question guys, ⎜Can I use my PC as SMTP server, I use Windows 7 . ⎜Already googled and Installed IIS ⎜but Still I can't send E-mail
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
Hi Ian, On Thu, 21 Dec 2023, Ian Jackson wrote: I have just become aware of #1058701 via the automated email that resulted from the removal of pm-utils. this is sad. The RM bug appeared on the tracker page of the package, in your packages overview, on the ftpmaster removals page (or on the bug page). It was also sent to debian-bugs-dist@lists.debian.org. Where would have been a better place to draw your attention to it? I am still using this package. I'm sure others are. The package *is* maintained in Debian. Hmm, the standards version is 3.9.7, the VCS URLs point to a no longer existing repository, the last upload was in 2019. This looks rather like an abandoned package. But of course, your mileage may vary I intend to re-upload the last version shortly (and reopen all the bug reports). Yes, please do so. Thorsten
Bug#1058701: pm-utils: unauthorised and uncommunicated removal
I have just become aware of #1058701 via the automated email that resulted from the removal of pm-utils. As the maintainer of this package, I do not agree with its removal. It has no RC bugs, is in testing, and is working software. I am still using this package. I'm sure others are. The package *is* maintained in Debian - I look for significant bugs and if there were any we would fix them. That the package hasn't seen much activity does not mean it doesn't work. I'm sure that it isn't normal Debian practice to remove a package from sid without consulting the maintainer. IMO the correct procedure if a package is thought to be unreleasable would be to remove it from testing first. [1] The submitter of #1058701 notes | Modern userspace uses systemd to perform suspend/resume instead. This should be a red flag, indicating that the removal was being requested by someone who thinks that Debian only supports systemd and that other more mature software for similar tasks ought to be removed. I intend to re-upload the last version shortly (and reopen all the bug reports). Thanks for your attention. DAM/CT: I am addressing this mail To you because the behaviour by the submitter of #1058701 represents behaviour I consider worthy of disciplinary action, and becauase it appears that there's been a process failure which allowed this removal to happen without anyone notifying the maintainer. It's probably best to deal with these matters in private. Ian. [1] That was attempted in #930869, which is unedifying. -- Ian JacksonThese opinions are my own. Pronouns: they/he. If I emailed you from @fyvzl.net or @evade.org.uk, that is a private address which bypasses my fierce spamfilter.
